Re: [Last-Call] vs Mobile IPv6 Last Call: <draft-ietf-6man-ra-pref64-07.txt> (Discovering PREF64 in Router Advertisements) to Proposed Standard - PLC consistency

[Alexandre Petrescu <[e-mail]>]

This is a diversion to Mobile IPv6 discussion, whose place is probably 
not on this last-call email list.

Le 08/11/2019 à 10:08, Ole Troan a écrit :
>> I'd have no way to know that RFC6275 recommends MaxRtrAdvInterval
>> value outside of the allowed range.
> 
> As long as you don't support Mobile IP do you care?
> And largely no-one supports mobile IP.
> Punted to transport.
> Like all the other hard problems at the network layer.

Ole,

I agree with you with the sense that Mobile IPv6 is largely not 
supported by anyone.

I like the briefness of the expression of being punted to transport 
layer like other hard problems.

However, I think the statement is too reducing.

When people implemented Mobile IPv6 and its derivatives they also 
implemented many add-ons that were needed for mobility such as driver 
support (software), WiFi handovers (link layer), NAT traversal and more. 
  Many of these aspects were migrated to 3GPP technologies, Windows 
technologies and more.  There are no RFCs for these technologies, but 
they were at least inspired, if not outright created, by works during 
Mobile IPv6 implementations.

In addition to the obvious transport layer that you mention, but that I 
dont know precisely what is referred to; in addition to the less obvious 
phy and mac layers replacements, Mobile IPv6 replacements also happen at 
application layer: maintaining connections up by a Mobile Router happen 
very often by using VPN software - I think that is app layer.  I am 
talking about deployed systems in trains and cars.

There is also the session layer: skype maintains reachability (but not 
ongoing calls) by resisting mobility events, at the session layer, I think.

So, I dont think one can reduce the Mobile IPv6 concepts to just being 
punted to transport layer.

And, as surprising as one might find, I think PREF64 is part of a set of 
technologies designed for mobility environments.  PREF64 solves a 
problem that DS-MIP (Dual Stack, RFC 5555) was supposed to solve too. 
And also a problem where 'double stack' (two stacks side by side, no 
translation) is supposed to be the best solution.

Yet, PREF64 seems to be more successful.  I do not disagree.

Alex

> O
>